Get data

import warnings
warnings.filterwarnings('ignore')

## Get data
from dataframes import final_df_dict

Average decision

from sympy import *
import matplotlib.pyplot as plt
import matplotlib as mpl
import numpy as np
from ipywidgets import Output
from tqdm import tqdm

df = final_df_dict['decision'].copy()
df.dropna(inplace=True)
df430_pre = df[(df['participant.inflation'] == '430') & (df['participant.day'] < 3)]
df1012_pre = df[(df['participant.inflation'] == '1012') & (df['participant.day'] < 3)]
df430_post = df[(df['participant.inflation'] == '430') & (df['participant.day'] >= 3)]
df1012_post = df[(df['participant.inflation'] == '1012') & (df['participant.day'] >= 3)]

dfs = {'430_pre': df430_pre,
       '1012_pre': df1012_pre,
       '430_post': df430_post,
       '1012_post': df1012_post}

## Calculate average value for each month
quantity_430_pre = np.array([])
quantity_1012_pre = np.array([])
quantity_430_post = np.array([])
quantity_1012_post = np.array([])

quantities = {}

for k, v in dfs.items():
    quantities[k] = np.array([])
    for n in range(1,121):
        quantities[k] = np.append(quantities[k], v[f'task.{n}.player.decision'].mean())



        # quantity_430_pre = np.append(quantity_430_pre, df430_pre[f'task.{n}.player.decision'].mean())
        # quantity_1012_pre = np.append(quantity_1012_pre, df1012_pre[f'task.{n}.player.decision'].mean())
        # quantity_430_post = np.append(quantity_430_post, df430_post[f'task.{n}.player.decision'].mean())
        # quantity_1012_post = np.append(quantity_1012_post, df1012_post[f'task.{n}.player.decision'].mean())

# qs = [quantity_430_pre,quantity_1012_pre,quantity_430_post,quantity_1012_post,]

# for i in range(len(qs)):
#     for n in range(1,121):
#         qs[i] = np.append(qs[i], dfs[i][f'task.{n}.player.decision'].mean())
# print('quantity', len(quantity_430_pre), type(quantity_430_pre))

months = np.array([n for n in range(1,121)])

for k, v in quantities.items():
    plt.errorbar(months, v, 0.3, fmt='o', label=k)
    
plt.legend(
    loc="upper right",
)

#  label=[
#     '430_pre',
#     '1012_pre',
#     '430_post',
#     '1012_post',

<matplotlib.legend.Legend at 0x24230a8e700>

Histograms of inflation change

df = final_df_dict['decision'].copy()
df.dropna(inplace=True)

cols430 = ['task.29.player.decision','task.30.player.decision','task.31.player.decision','task.32.player.decision',
              'task.59.player.decision','task.60.player.decision','task.61.player.decision','task.62.player.decision',
              'task.89.player.decision','task.90.player.decision','task.91.player.decision','task.92.player.decision',]
cols1012 = ['task.11.player.decision','task.12.player.decision','task.13.player.decision','task.14.player.decision',
              'task.35.player.decision','task.36.player.decision','task.37.player.decision','task.38.player.decision',
              'task.59.player.decision','task.60.player.decision','task.61.player.decision','task.62.player.decision',
              'task.83.player.decision','task.84.player.decision','task.85.player.decision','task.86.player.decision',
              'task.107.player.decision','task.108.player.decision','task.109.player.decision','task.110.player.decision',]

df430_pre = df[cols430][(df['participant.inflation'] == '430') & (df['participant.day'] < 3)]
df1012_pre = df[cols1012][(df['participant.inflation'] == '1012') & (df['participant.day'] < 3)]
df430_post = df[cols430][(df['participant.inflation'] == '430') & (df['participant.day'] >= 3)]
df1012_post = df[cols1012][(df['participant.inflation'] == '1012') & (df['participant.day'] >= 3)]

dfs = {'430_pre': df430_pre,
       '1012_pre': df1012_pre,
       '430_post': df430_post,
       '1012_post': df1012_post}

# for k, v in dfs.items():
#     plt.hist(v)

# plt.hist(df430_pre, df430_post)

# df430_post


plt.hist([df430_pre[cols430[1+1]],df430_post[cols430[1+1]],df430_pre[cols430[2+1]],
          df430_post[cols430[2+1]]], alpha=0.5, label=['430_pre 31','430_post 31', 
                                                     '430_pre 32','430_post 32'])
# plt.bar_label([df430_pre,df430_post], fontsize=20, color='navy')
# plt.hist(df430_post, alpha=0.5, label='430_post')
plt.legend(loc='upper right')
plt.show()

plt.hist([df430_pre[cols430[5+1]],df430_post[cols430[5+1]],df430_pre[cols430[6+1]],
          df430_post[cols430[6+1]]], alpha=0.5, label=['430_pre 61','430_post 61', 
                                                     '430_pre 62','430_post 62'])
# plt.bar_label([df430_pre,df430_post], fontsize=20, color='navy')
# plt.hist(df430_post, alpha=0.5, label='430_post')
plt.legend(loc='upper right')
plt.show()

plt.hist([df430_pre[cols430[10]],df430_post[cols430[10]],df430_pre[cols430[11]],
          df430_post[cols430[11]]], alpha=0.5, label=['430_pre 91','430_post 91', 
                                                     '430_pre 92','430_post 92'])
# plt.bar_label([df430_pre,df430_post], fontsize=20, color='navy')
# plt.hist(df430_post, alpha=0.5, label='430_post')
plt.legend(loc='upper right')
plt.show()


plt.hist([df1012_pre[cols1012[1]],df1012_post[cols1012[1]],df1012_pre[cols1012[2]],
          df1012_post[cols1012[2]]], alpha=0.5, label=['1012_pre 12','1012_post 12', 
                                                     '1012_pre 13','1012_post 13'])
plt.legend(loc='upper right')
plt.show()

(442, 126)
(442, 126)
(117, 12) (115, 20) (105, 12) (105, 20)

Inflation perceptions

est = final_df_dict['inf_estimate'].copy()
inf_1012 = [0.45, 60.79, 0.45, 60.79, 0.45, 60.79, 0.45, 60.79, 0.45, 60.79]
inf_430 = [0.38, 0.47, 26.85, 55.49, 64.18, 0.38, 0.47, 26.85, 55.49, 64.18]
rev_1012 = [x[1] - x[0] for x in zip(inf_1012, inf_1012[1:])]
rev_430 = [x[1] - x[0] for x in zip(inf_430, inf_430[1:])]

for n in range(2,11):
    est[f'{12 * n}-{12 * (n - 1)}'] = est[f'task.{n * 12}.player.inf_estimate'] -\
    est[f'task.{(n - 1) * 12}.player.inf_estimate']

est['revisions'] = np.nan

for i in range(len(est)):
    x = 0
    for n in range(2,11):
        if est[f'{12 * n}-{12 * (n - 1)}'].iloc[i] != 0:
            x += 1
        else:
            pass
    est['revisions'].iloc[i] = x
    
## Graph estimates
cols = [c for c in est.columns.to_list() if ('task.' in c)]

fig, ax = plt.subplots(figsize=(24, 10))
## Actual inflation
## 10x12 pre
df = est[cols][
    (est['participant.inflation'] == '1012') & (est['participant.day'] < 3) & (
        est['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '10x12 pre'
df.columns = df_cols
df.plot(ax=ax, color='r', alpha=0.1)
# ax.get_legend().remove()
## 10x12 post
df = est[cols][
    (est['participant.inflation'] == '1012') & (est['participant.day'] >= 3) & (
        est['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '10x12 post'
df.columns = df_cols
df.plot(ax=ax, color='b', alpha=0.1)
# ax.get_legend().remove()
plt.axhline(y=60.79, color='g', linestyle=':', label='max inflation')
plt.axhline(y=0, color='black', linestyle='-')
plt.plot(inf_1012, color='#f2f542', linewidth=4, label='actual inflation')
plt.legend(loc = 'upper left')
plt.title('10x12 perceptions')


fig, ax = plt.subplots(figsize=(24, 10))
## 4x30 pre
df = est[cols][
    (est['participant.inflation'] == '430') & (est['participant.day'] < 3) & (
        est['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '4x30 pre'
df.columns = df_cols
df.plot(ax=ax, color='r', alpha=0.1)
# ax.get_legend().remove()

## 4x30 post
df = est[cols][
    (est['participant.inflation'] == '430') & (est['participant.day'] >= 3) & (
        est['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '4x30 post'
df.columns = df_cols
df.plot(ax=ax, color='b', alpha=0.1)
# ax.get_legend().remove()
plt.axhline(y=64.18, color='g', linestyle=':', label='max inflation')
plt.axhline(y=0, color='black', linestyle='-')
plt.plot(inf_430, color='#f2f542', linewidth=4, label='actual inflation')
plt.legend(loc = 'upper left')
plt.title('4x30 perceptions')


## Graph revisions
cols = [c for c in est.columns.to_list() if ('-' in c)]
fig, ax = plt.subplots(figsize=(24, 10))

## 10x12 pre
df = est[cols][
    (est['participant.inflation'] == '1012') & (est['participant.day'] < 3) & (
        est['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '10x12  pre'
df.columns = df_cols
df.plot(ax=ax, color='r', alpha=0.1)
# ax.get_legend().remove()

## 10x12 post
df = est[cols][
    (est['participant.inflation'] == '1012') & (est['participant.day'] >= 3) & (
        est['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '10x12  post'
df.columns = df_cols
df.plot(ax=ax, color='b', alpha=0.1)
# ax.get_legend().remove()
plt.axhline(y=60.79, color='g', linestyle=':', label='max inflation')
plt.axhline(y=0, color='black', linestyle='-')
plt.plot(rev_1012, color='#f2f542', linewidth=4, label='correct revision')
plt.legend(loc = 'upper left')
plt.title('10x12 revisions')

fig, ax = plt.subplots(figsize=(24, 10))

## 4x30 pre
df = est[cols][
    (est['participant.inflation'] == '430') & (est['participant.day'] < 3) & (
        est['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '4x30  pre'
df.columns = df_cols
df.plot(ax=ax, color='r', alpha=0.1)
# ax.get_legend().remove()

## 4x30 post
df = est[cols][
    (est['participant.inflation'] == '430') & (est['participant.day'] >= 3) & (
        est['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '4x30  post'
df.columns = df_cols
df.plot(ax=ax, color='b', alpha=0.1)
# ax.get_legend().remove()
plt.axhline(y=64.18, color='g', linestyle=':', label='max inflation')
plt.axhline(y=0, color='black', linestyle='-')
plt.plot(rev_430, color='#f2f542', linewidth=4, label='correct revision')
plt.legend(loc = 'upper left')
plt.title('4x30 revisions')

Text(0.5, 1.0, '4x30 revisions')

FFT

def fft_row_magnitudes(row):
    fft = np.fft.fft(row)
    magnitudes = np.abs(fft)
    # power_spectrum = np.abs(fft)**2

est['estimate'] = ''

## Create columns for lists of estimates
for i in range(len(est)):
    estimate = est.iloc[i].to_list()
    estimate = estimate[6:15]

    # ## Convert stock values to int
    # stock = [int(i) for i in stock]
    est['estimate'].iloc[i] = estimate

est['fft_magnitudes'] = ''

for i in range(len(est)):
    fft = np.fft.fft(est['estimate'].iloc[i])
    est['fft_magnitudes'].iloc[i] = np.abs(fft)
    # power_spectrum = np.abs(fft)**2

# est['fft_magnitudes'] = est['estimate'].apply(fft_row_magnitudes)

## Frequencies of each curve
N = len(est['fft_magnitudes'].iloc[1])  # length of the magnitude list
sampling_rate = 1  # assuming a sampling rate of 1
frequencies = [(i / N) * sampling_rate for i in range(N)]
periodicities = [1 / i for i in frequencies if i != 0]
print(frequencies, len(frequencies))
print(periodicities, len(frequencies))
N = len(est['fft_magnitudes'].iloc[2])  # length of the magnitude list
sampling_rate = 1  # assuming a sampling rate of 1
frequencies = [(i / N) * sampling_rate for i in range(N)]
periodicities = [1 / i for i in frequencies if i != 0]
print(frequencies, len(frequencies))
print(periodicities, len(frequencies))

[0.0, 0.1111111111111111, 0.2222222222222222, 0.3333333333333333, 0.4444444444444444, 0.5555555555555556, 0.6666666666666666, 0.7777777777777778, 0.8888888888888888] 9
[9.0, 4.5, 3.0, 2.25, 1.7999999999999998, 1.5, 1.2857142857142856, 1.125] 9
[0.0, 0.1111111111111111, 0.2222222222222222, 0.3333333333333333, 0.4444444444444444, 0.5555555555555556, 0.6666666666666666, 0.7777777777777778, 0.8888888888888888] 9
[9.0, 4.5, 3.0, 2.25, 1.7999999999999998, 1.5, 1.2857142857142856, 1.125] 9

Histograms

cols = [c for c in est.columns.to_list() if ('task.' in c)]

fig, ax = plt.subplots(figsize=(20,20))
df1 = est[cols][(est['participant.inflation'] == '430') & (
    est['participant.day'] < 3
)]
df1.hist(ax=ax, color='r', label='pre')

fig, ax = plt.subplots(figsize=(20,20))
# ax = fig.gca()
df2 = est[cols][(est['participant.inflation'] == '430') & (
    est['participant.day'] >= 3
)]
df2.hist(ax=ax, color='b', label='post')
# ax.legend(loc='upper right')
# plt.show()


fig, ax = plt.subplots(figsize=(20,20))
df1 = est[cols][(est['participant.inflation'] == '1012') & (
    est['participant.day'] < 3
)]
df1.hist(ax=ax, color='r', label='pre')

fig, ax = plt.subplots(figsize=(20,20))
# ax = fig.gca()
df2 = est[cols][(est['participant.inflation'] == '1012') & (
    est['participant.day'] >= 3
)]
df2.hist(ax=ax, color='b', label='post')

array([[<AxesSubplot:title={'center':'task.12.player.inf_estimate'}>,
        <AxesSubplot:title={'center':'task.24.player.inf_estimate'}>,
        <AxesSubplot:title={'center':'task.36.player.inf_estimate'}>],
       [<AxesSubplot:title={'center':'task.48.player.inf_estimate'}>,
        <AxesSubplot:title={'center':'task.60.player.inf_estimate'}>,
        <AxesSubplot:title={'center':'task.72.player.inf_estimate'}>],
       [<AxesSubplot:title={'center':'task.84.player.inf_estimate'}>,
        <AxesSubplot:title={'center':'task.96.player.inf_estimate'}>,
        <AxesSubplot:title={'center':'task.108.player.inf_estimate'}>],
       [<AxesSubplot:title={'center':'task.120.player.inf_estimate'}>,
        <AxesSubplot:>, <AxesSubplot:>]], dtype=object)

Inflation expectations

exp = final_df_dict['inf_expectation'].copy()
inf_1012 = [60.79, 0.45, 60.79, 0.45, 60.79, 0.45, 60.79, 0.45, 60.79]
inf_430 = [0.47, 26.85, 55.49, 64.18, 0.38, 0.47, 26.85, 55.49, 64.18]
rev_1012 = [x[1] - x[0] for x in zip(inf_1012, inf_1012[1:])]
rev_430 = [x[1] - x[0] for x in zip(inf_430, inf_430[1:])]

for n in range(2,10):
    exp[f'{12 * n}-{12 * (n - 1)}'] = exp[f'task.{n * 12}.player.inf_expectation'] -\
    exp[f'task.{(n - 1) * 12}.player.inf_expectation']

exp['revisions'] = np.nan

for i in range(len(exp)):
    x = 0
    for n in range(2,10):
        if exp[f'{12 * n}-{12 * (n - 1)}'].iloc[i] != 0:
            x += 1
        else:
            pass
    exp['revisions'].iloc[i] = x
    
## Graph expimates
cols = [c for c in exp.columns.to_list() if ('task.' in c)]

fig, ax = plt.subplots(figsize=(24, 10))
## Actual inflation
## 10x12 pre
df = exp[cols][
    (exp['participant.inflation'] == '1012') & (exp['participant.day'] < 3) & (
        exp['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '10x12 pre'
df.columns = df_cols
df.plot(ax=ax, color='r', alpha=0.1)
# ax.get_legend().remove()
## 10x12 post
df = exp[cols][
    (exp['participant.inflation'] == '1012') & (exp['participant.day'] >= 3) & (
        exp['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '10x12 post'
df.columns = df_cols
df.plot(ax=ax, color='b', alpha=0.1)
# ax.get_legend().remove()
plt.axhline(y=60.79, color='g', linestyle=':', label='max inflation')
plt.axhline(y=0, color='black', linestyle='-')
plt.plot(inf_1012, color='#f2f542', linewidth=4, label='actual inflation')
plt.legend(loc = 'upper left')
plt.title('10x12 expectations')


fig, ax = plt.subplots(figsize=(24, 10))
## 4x30 pre
df = exp[cols][
    (exp['participant.inflation'] == '430') & (exp['participant.day'] < 3) & (
        exp['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '4x30 pre'
df.columns = df_cols
df.plot(ax=ax, color='r', alpha=0.1)
# ax.get_legend().remove()

## 4x30 post
df = exp[cols][
    (exp['participant.inflation'] == '430') & (exp['participant.day'] >= 3) & (
        exp['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '4x30 post'
df.columns = df_cols
df.plot(ax=ax, color='b', alpha=0.1)
# ax.get_legend().remove()
plt.axhline(y=64.18, color='g', linestyle=':', label='max inflation')
plt.axhline(y=0, color='black', linestyle='-')
plt.plot(inf_430, color='#f2f542', linewidth=4, label='actual inflation')
plt.legend(loc = 'upper left')
plt.title('4x30 expectations')


## Graph revisions
cols = [c for c in exp.columns.to_list() if ('-' in c)]
fig, ax = plt.subplots(figsize=(24, 10))

## 10x12 pre
df = exp[cols][
    (exp['participant.inflation'] == '1012') & (exp['participant.day'] < 3) & (
        exp['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '10x12  pre'
df.columns = df_cols
df.plot(ax=ax, color='r', alpha=0.1)
# ax.get_legend().remove()

## 10x12 post
df = exp[cols][
    (exp['participant.inflation'] == '1012') & (exp['participant.day'] >= 3) & (
        exp['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '10x12  post'
df.columns = df_cols
df.plot(ax=ax, color='b', alpha=0.1)
# ax.get_legend().remove()
plt.axhline(y=60.79, color='g', linestyle=':', label='max inflation')
plt.axhline(y=0, color='black', linestyle='-')
plt.plot(rev_1012, color='#f2f542', linewidth=4, label='actual inflation')
plt.legend(loc = 'upper left')
plt.title('10x12 revisions')

fig, ax = plt.subplots(figsize=(24, 10))

## 4x30 pre
df = exp[cols][
    (exp['participant.inflation'] == '430') & (exp['participant.day'] < 3) & (
        exp['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '4x30  pre'
df.columns = df_cols
df.plot(ax=ax, color='r', alpha=0.1)
# ax.get_legend().remove()

## 4x30 post
df = exp[cols][
    (exp['participant.inflation'] == '430') & (exp['participant.day'] >= 3) & (
        exp['revisions'] >= 0
    )
].T
df_cols = ['_' + str(col) for col in df.columns]
df_cols[0] = '4x30  post'
df.columns = df_cols
df.plot(ax=ax, color='b', alpha=0.1)
# ax.get_legend().remove()
plt.axhline(y=64.18, color='g', linestyle=':', label='max inflation')
plt.axhline(y=0, color='black', linestyle='-')
plt.plot(rev_430, color='#f2f542', linewidth=4, label='actual inflation')
plt.legend(loc = 'upper left')
plt.title('4x30 revisions')

Text(0.5, 1.0, '4x30 revisions')

Histograms

cols = [c for c in exp.columns.to_list() if ('task.' in c)]

fig, ax = plt.subplots(figsize=(20,20))
df1 = exp[cols][(exp['participant.inflation'] == '430') & (
    exp['participant.day'] < 3
)]
df1.hist(ax=ax, color='r', label='pre')

fig, ax = plt.subplots(figsize=(20,20))
# ax = fig.gca()
df2 = exp[cols][(exp['participant.inflation'] == '430') & (
    exp['participant.day'] >= 3
)]
df2.hist(ax=ax, color='b', label='post')
# ax.legend(loc='upper right')
# plt.show()


fig, ax = plt.subplots(figsize=(20,20))
df1 = exp[cols][(exp['participant.inflation'] == '1012') & (
    exp['participant.day'] < 3
)]
df1.hist(ax=ax, color='r', label='pre')

fig, ax = plt.subplots(figsize=(20,20))
# ax = fig.gca()
df2 = exp[cols][(exp['participant.inflation'] == '1012') & (
    exp['participant.day'] >= 3
)]
df2.hist(ax=ax, color='b', label='post')

array([[<AxesSubplot:title={'center':'task.12.player.inf_expectation'}>,
        <AxesSubplot:title={'center':'task.24.player.inf_expectation'}>,
        <AxesSubplot:title={'center':'task.36.player.inf_expectation'}>],
       [<AxesSubplot:title={'center':'task.48.player.inf_expectation'}>,
        <AxesSubplot:title={'center':'task.60.player.inf_expectation'}>,
        <AxesSubplot:title={'center':'task.72.player.inf_expectation'}>],
       [<AxesSubplot:title={'center':'task.84.player.inf_expectation'}>,
        <AxesSubplot:title={'center':'task.96.player.inf_expectation'}>,
        <AxesSubplot:title={'center':'task.108.player.inf_expectation'}>]],
      dtype=object)

Estimate revisions

plt.subplots()
est['revisions'][(est['participant.inflation'] == '1012') & (
    est['participant.day'] < 3
)].hist(color='r', label='10x12 pre-intervention')
est['revisions'][(est['participant.inflation'] == '1012') & (
    est['participant.day'] >= 3
)].hist(color='b', alpha = 0.4, label='10x12 post-intervention')
plt.legend(loc='upper left')

plt.subplots()
est['revisions'][(est['participant.inflation'] == '430') & (
    est['participant.day'] < 3
)].hist(color='r', label='4x30 pre-intervention')
est['revisions'][(est['participant.inflation'] == '430') & (
    est['participant.day'] >= 3
)].hist(color='b', alpha = 0.4, label='4x30 post-intervention')
plt.legend(loc='upper left')

<matplotlib.legend.Legend at 0x24229b8d910>

Descriptive statistics of estimate revisions

cols = [c for c in est.columns.to_list() if ('task.' not in c) and ('.day' not in c)]
est[cols].groupby(['participant.inflation', est['participant.day'] < 3]).describe().T

	participant.inflation	1012		430
	participant.day	False	True	False	True
24-12	count	105.000000	113.000000	105.000000	112.000000
	mean	18.857143	18.212389	0.219048	1.598214
	std	23.566774	23.042837	6.062148	15.138674
	min	-23.000000	-67.000000	-37.000000	-48.000000
	25%	2.000000	4.000000	0.000000	0.000000
	50%	8.000000	9.000000	0.000000	0.000000
	75%	39.000000	37.000000	2.000000	2.000000
	max	83.000000	78.000000	24.000000	97.000000
36-24	count	105.000000	113.000000	105.000000	112.000000
	mean	-12.704762	-10.185841	7.933333	8.660714
	std	26.516658	24.050374	10.560509	14.832733
	min	-83.000000	-74.000000	-16.000000	-82.000000
	25%	-23.000000	-20.000000	1.000000	1.000000
	50%	-2.000000	-2.000000	4.000000	6.000000
	75%	2.000000	4.000000	13.000000	15.000000
	max	46.000000	54.000000	62.000000	64.000000
48-36	count	105.000000	113.000000	105.000000	112.000000
	mean	15.380952	14.761062	8.361905	6.464286
	std	22.322109	24.226341	15.699169	19.043775
	min	-21.000000	-54.000000	-47.000000	-137.000000
	25%	1.000000	1.000000	1.000000	0.000000
	50%	5.000000	7.000000	4.000000	5.000000
	75%	20.000000	27.000000	17.000000	12.250000
	max	67.000000	95.000000	46.000000	85.000000
60-48	count	105.000000	113.000000	105.000000	112.000000
	mean	-12.895238	-9.398230	6.466667	10.776786
	std	25.139581	26.335588	18.812877	23.533648
	min	-66.000000	-80.000000	-58.000000	-59.000000
	25%	-20.000000	-20.000000	-1.000000	0.000000
	50%	-4.000000	-3.000000	2.000000	4.000000
	75%	1.000000	2.000000	10.000000	15.250000
	max	50.000000	65.000000	91.000000	159.000000
72-60	count	105.000000	113.000000	105.000000	112.000000
	mean	14.333333	11.548673	-14.704762	-14.035714
	std	22.759050	24.433946	24.297406	28.418979
	min	-38.000000	-50.000000	-99.000000	-96.000000
	25%	1.000000	-1.000000	-21.000000	-24.250000
	50%	6.000000	5.000000	-5.000000	-6.500000
	75%	21.000000	21.000000	0.000000	1.250000
	max	70.000000	99.000000	19.000000	59.000000
84-72	count	105.000000	113.000000	105.000000	112.000000
	mean	-12.323810	-12.309735	0.342857	-1.419643
	std	24.349418	23.700919	9.212062	20.052920
	min	-73.000000	-98.000000	-45.000000	-161.000000
	25%	-20.000000	-23.000000	-1.000000	-4.250000
	50%	-3.000000	-4.000000	0.000000	0.000000
	75%	1.000000	2.000000	2.000000	3.000000
	max	49.000000	34.000000	35.000000	46.000000
96-84	count	105.000000	112.000000	105.000000	112.000000
	mean	18.485714	18.205357	6.009524	6.455357
	std	25.223659	25.981330	15.573335	15.659029
	min	-25.000000	-27.000000	-90.000000	-53.000000
	25%	0.000000	0.000000	0.000000	0.000000
	50%	8.000000	10.500000	2.000000	4.500000
	75%	31.000000	29.250000	13.000000	14.000000
	max	99.000000	114.000000	39.000000	69.000000
108-96	count	105.000000	112.000000	105.000000	112.000000
	mean	-13.714286	-13.232143	11.161905	10.580357
	std	27.040822	24.782652	19.730942	27.504181
	min	-95.000000	-74.000000	-32.000000	-185.000000
	25%	-28.000000	-26.000000	0.000000	0.000000
	50%	-3.000000	-5.000000	5.000000	8.000000
	75%	3.000000	1.000000	20.000000	17.250000
	max	48.000000	50.000000	91.000000	121.000000
120-108	count	105.000000	112.000000	105.000000	112.000000
	mean	29.009524	25.250000	16.600000	18.125000
	std	28.919815	28.640204	23.212066	26.589683
	min	-49.000000	-26.000000	-25.000000	-47.000000
	25%	8.000000	2.750000	0.000000	2.750000
	50%	24.000000	14.500000	10.000000	12.000000
	75%	47.000000	41.250000	24.000000	28.250000
	max	99.000000	99.000000	94.000000	174.000000
revisions	count	105.000000	115.000000	105.000000	117.000000
	mean	8.371429	8.391304	7.619048	7.777778
	std	1.475812	1.309182	1.589515	1.509231
	min	0.000000	1.000000	0.000000	0.000000
	25%	8.000000	8.000000	7.000000	7.000000
	50%	9.000000	9.000000	8.000000	8.000000
	75%	9.000000	9.000000	9.000000	9.000000
	max	9.000000	9.000000	9.000000	9.000000

Expectations revisions

plt.subplots()
exp['revisions'][(exp['participant.inflation'] == '1012') & (
    exp['participant.day'] < 3
)].hist(color='r', label='10x12 pre-intervention')
exp['revisions'][(exp['participant.inflation'] == '1012') & (
    exp['participant.day'] >= 3
)].hist(color='b', alpha = 0.4, label='10x12 post-intervention')
plt.legend(loc='upper left')

plt.subplots()
exp['revisions'][(exp['participant.inflation'] == '430') & (
    exp['participant.day'] < 3
)].hist(color='r', label='4x30 pre-intervention')
exp['revisions'][(exp['participant.inflation'] == '430') & (
    exp['participant.day'] >= 3
)].hist(color='b', alpha = 0.4, label='4x30 post-intervention')
plt.legend(loc='upper left')

<matplotlib.legend.Legend at 0x2422f4f1730>

Descriptive statistics of expectations revisions

cols = [c for c in exp.columns.to_list() if ('task.' not in c) and ('.day' not in c)]
exp[cols].groupby(['participant.inflation', exp['participant.day'] < 3]).describe().T

	participant.inflation	1012		430
	participant.day	False	True	False	True
24-12	count	105.000000	113.000000	105.000000	112.000000
	mean	4.914286	16.752212	2.095238	-1.026786
	std	25.220228	22.193762	10.407582	15.098750
	min	-105.000000	-27.000000	-24.000000	-45.000000
	25%	-3.000000	2.000000	0.000000	-4.250000
	50%	3.000000	8.000000	1.000000	0.000000
	75%	12.000000	26.000000	4.000000	2.000000
	max	89.000000	95.000000	57.000000	106.000000
36-24	count	105.000000	113.000000	105.000000	112.000000
	mean	4.961905	-3.752212	2.990476	9.857143
	std	22.217857	21.998186	16.688605	15.200674
	min	-59.000000	-110.000000	-66.000000	-35.000000
	25%	-1.000000	-11.000000	-1.000000	0.750000
	50%	1.000000	0.000000	2.000000	7.000000
	75%	12.000000	6.000000	9.000000	17.000000
	max	106.000000	61.000000	66.000000	68.000000
48-36	count	105.000000	113.000000	105.000000	112.000000
	mean	-1.542857	2.955752	2.247619	2.178571
	std	20.794336	16.102013	14.671913	15.858163
	min	-66.000000	-49.000000	-43.000000	-75.000000
	25%	-7.000000	-3.000000	-1.000000	-4.250000
	50%	1.000000	3.000000	1.000000	3.000000
	75%	8.000000	12.000000	9.000000	12.000000
	max	74.000000	55.000000	57.000000	36.000000
60-48	count	105.000000	113.000000	105.000000	112.000000
	mean	4.019048	2.522124	5.028571	8.017857
	std	23.030490	22.377483	19.515505	20.558413
	min	-69.000000	-72.000000	-71.000000	-96.000000
	25%	-5.000000	-8.000000	-1.000000	-1.250000
	50%	0.000000	0.000000	3.000000	3.000000
	75%	10.000000	8.000000	10.000000	17.250000
	max	71.000000	100.000000	82.000000	65.000000
72-60	count	105.000000	113.000000	105.000000	112.000000
	mean	-1.904762	-1.646018	-4.152381	-5.517857
	std	19.376097	21.904387	19.614938	20.719483
	min	-66.000000	-77.000000	-93.000000	-78.000000
	25%	-7.000000	-7.000000	-9.000000	-15.000000
	50%	1.000000	0.000000	-2.000000	-1.500000
	75%	8.000000	5.000000	2.000000	4.500000
	max	57.000000	65.000000	62.000000	42.000000
84-72	count	105.000000	113.000000	105.000000	112.000000
	mean	5.666667	3.415929	1.980952	-1.901786
	std	19.741925	21.775939	15.160666	16.515873
	min	-29.000000	-58.000000	-50.000000	-54.000000
	25%	-5.000000	-4.000000	-4.000000	-8.000000
	50%	0.000000	1.000000	0.000000	0.000000
	75%	9.000000	10.000000	5.000000	5.000000
	max	66.000000	80.000000	58.000000	45.000000
96-84	count	105.000000	112.000000	105.000000	112.000000
	mean	-3.476190	-1.839286	0.314286	3.267857
	std	24.493365	26.953598	16.731657	17.780434
	min	-67.000000	-94.000000	-84.000000	-63.000000
	25%	-7.000000	-13.250000	-3.000000	-2.000000
	50%	0.000000	0.000000	1.000000	3.000000
	75%	6.000000	8.250000	6.000000	12.000000
	max	99.000000	63.000000	45.000000	69.000000
108-96	count	105.000000	112.000000	105.000000	112.000000
	mean	8.419048	4.616071	8.380952	5.696429
	std	25.232298	24.659938	16.802233	26.788070
	min	-95.000000	-66.000000	-28.000000	-183.000000
	25%	-2.000000	-5.000000	-1.000000	-1.250000
	50%	3.000000	2.000000	5.000000	1.500000
	75%	16.000000	11.000000	13.000000	14.250000
	max	70.000000	99.000000	85.000000	72.000000
revisions	count	105.000000	115.000000	105.000000	117.000000
	mean	7.171429	7.391304	6.942857	7.094017
	std	1.369005	1.226145	1.343278	1.383293
	min	0.000000	2.000000	0.000000	1.000000
	25%	7.000000	7.000000	7.000000	7.000000
	50%	8.000000	8.000000	7.000000	8.000000
	75%	8.000000	8.000000	8.000000	8.000000
	max	8.000000	8.000000	8.000000	8.000000

Decisions distributions

df = final_df_dict['decision'].copy()

cols = [c for c in df.columns.to_list() if 'task.' in c]

df[cols] = df[cols].astype(float)

fig, ax = plt.subplots(figsize=(30,30))
df1 = df[cols][(df['participant.inflation'] == '1012') & (
    df['participant.day'] < 3)]
df1.hist(ax=ax, bins=20, color='r', label='pre')
plt.legend(loc='upper left')

fig, ax = plt.subplots(figsize=(30,30))
df1 = df[cols][(df['participant.inflation'] == '1012') & (
    df['participant.day'] >= 3)]
df1.hist(ax=ax, bins=20, color='b', label='post')
plt.legend(loc='upper left')

fig, ax = plt.subplots(figsize=(30,30))
df1 = df[cols][(df['participant.inflation'] == '430') & (
    df['participant.day'] < 3)]
df1.hist(ax=ax, bins=20, color='r', label='pre')
plt.legend(loc='upper left')

fig, ax = plt.subplots(figsize=(30,30))
df1 = df[cols][(df['participant.inflation'] == '430') & (
    df['participant.day'] >= 3)]
df1.hist(ax=ax, bins=20, color='b', label='post')
plt.legend(loc='upper left')

No handles with labels found to put in legend.
No handles with labels found to put in legend.
No handles with labels found to put in legend.
No handles with labels found to put in legend.

<matplotlib.legend.Legend at 0x2422e88cc40>

End